Research on new multi-layer motion features effective for video recognition
| Project/Area Number |
15K00249
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Meisei University (2017)
Waseda University (2015-2016) |
|---|
Principal Investigator |
Ueki Kazuya 明星大学, 情報学部, 准教授 (80580638)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2017: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2015: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
|
|---|
| Keywords | 映像認識 / 映像検索 / モーション認識 / 多層 / CNN / TRECVID / モーション特徴 / 識別的モーション特徴 / オプティカルフロー画像 |
|---|
| Outline of Final Research Achievements |
With regard to the extraction of discriminative and highly accurate motion features with multi-layered structure, we were able to construct video retrieval system ahead of schedule by extracting the gradient motion features, specifically the optical flow features, and training them with convolutional neural networks. In addition, we evaluated our system with a large-scale video database at the TREC Video Retrieval Evaluation benchmark (TRECVID) organized by the National Institute of Standards and Technology and confirmed its effectiveness.
|
Report
(4 results)
Research Products
(10 results)
